Abstract Intro: Conventional training methods are intended to teach and offer the most human way of teaching but is it the most effective and can another method improve or eliminate nonvalue adding activities. This master thesis tackles the difference between traditional training versus AR supported training, studying different aspects of training to see the advantages and disadvantages of the methods. The research questions were the following:  RQ1: Is the learning process improved with or without AR supported technology?   RQ2: What are the benefits of AR use in industry?  RQ3: Will the investment of AR implementation in the learning process pay off?  Method: To answer the research questions, a case study at a case company was conducted. The case study consisted of working with two companies, an external company for a theoretical answer based on experiences and an experimental study at the internal case company. In parallel to this case study a literature review was done to answer the research questions.  Theory and Literature Review: Literature for the frame of reference was gathered to support the thesis and give the perspectives of the investigated area. The literature review sought to find answers to the research questions and the search was systematically formed to focus on data gathering to answer the questions.  Analyses: The findings from this research shows that AR supported training is beneficial and has the potential to eliminate nonvalue adding activities. AR has many capabilities and research shows that AR in general has positive impact.  Conclusion: AR-applied training has both advantages and disadvantages, but the potential of improvements is high. It is difficult to conclude if it is economical to invest in AR, with the advances the technology holds on to now since the outcome can depend on several factors from different parameters. However, AR does contribute to more effective learning and further benefits within the applied area. It has been proven to reduce the error rate and thus may increase the quality of the product.

Industry 4.0 is a collective name for several technological innovations that, when combined, among other things, provide an exponential potential for increased operational excellence in manufacturing. This thesis digs down into which technologies that are relevant in the context of predictive maintenance and how these can be integrated into existing theory in order to create value through increased e↵ectiveness. The primary findings can be condensed down into one general principal - uniformity. In order to leverage industry 4.0, and through it achieve a higher level of automatization, all data flow must be as canonical as possible. This is what allows both for bi-directional communication at scale, and higher-level decisionmaking algorithms to be deployed over a wide range of hardware. / Industri 4.0 är ett samlingsnamn för ett flertal tekniska innovationer vilka, tillsammans, möjliggör en potentiell förbättring av operational excellence som ökar exponentiellt mot antalet aopterade teknologier. Detta arbete dyker ned i vilka teknologier som skapar mest värde i kontexten predictive maintenance. Arbetet studerar även existerande orginatorisk teor och hur dessa kan slås samman. Det primära resultatet kan summeras som att fokus bör ligga på en canonical model för den data som genereras, och skickas ned till maskiner på fabriksgolvet. Uniform data spelar även en nyckelroll i att facilitera för beslutsfattande algoritmer då dessa annars enbat skulle gå att applicera på specifika maskiner.

Planning and management of logistics and material flow are widely studied and two key factors contributing to company competitiveness. Automation in material flow is recognized in efficient and profitable factories in the context of today’s smart industry, however, the operators are playing a significant role as well. The purpose of the study was to identify which criteria could be used to determine the level of automation in a material flow of the industrial factory. A model was developed to practically support the decision making on the level of automation for the case company that was going to build a pilot line for battery manufacturing. The question for the case company was to decide whether the process should be fully automated, manual, or semi-automated in its trial production to avoid costly reconfigurations when a full production starts. In this study a literature review was conducted in the form of previous research to describe which criteria were important to decide the level of automation. The literature study together with site visiting, interviews, survey and document analysis was used for the formulation of research questions, establishment of methodology and model development. The study has identified different criteria whereas six of them are shared by both literature studies and the case company. Two unique criteria were identified in the case company but not found in the literature: lead time and recruitment. The lead time refers to the time from planning to finished process and the recruitment is about how difficult to recruit people with the right skills. Two theoretical contributions presented in this study: new criteria when deciding on the level of automation and a model that assists in decisions regarding the level of automation in early phases. The model is also a practical contribution to the case company. Finally, some suggestions for further research within the area are presented.

The smart factory is at the heart of Industry 4.0 and is the new paradigm for establishing advanced manufacturing systems and realizing modern manufacturing objectives such as mass customization, automation, efficiency, and self-organization all at once. Such manufacturing systems, however, are characterized by dynamic and complex environments where a large number of decisions should be made for smart components such as production machines and the material handling system in a real-time and optimal manner. AI offers key intelligent control approaches in order to realize efficiency, agility, and automation all at once. One of the most challenging problems faced in this regard is uncertainty, meaning that due to the dynamic nature of the smart manufacturing environments, sudden seen or unseen events occur that should be handled in real-time. Due to the complexity and high-dimensionality of smart factories, it is not possible to predict all the possible events or prepare appropriate scenarios to respond. Reinforcement learning is an AI technique that provides the intelligent control processes needed to deal with such uncertainties. Due to the distributed nature of smart factories and the presence of multiple decision-making components, multi-agent reinforcement learning (MARL) should be incorporated instead of single-agent reinforcement learning (SARL), which, due to the complexities involved in the development process, has attracted less attention. In this research, we will review the literature on the applications of MARL to tasks within a smart factory and then demonstrate a mapping connecting smart factory attributes to the equivalent MARL features, based on which we suggest MARL to be one of the most effective approaches for implementing the control mechanism for smart factories.

The industrial company Scania CV AB is a world leader in the manufacturing of commercial vehicles. They offer a modular systems that include heavy trucks and buses that can be configured to a range of different needs. However, this adaptability leads to a problem where each order can have a large variance of assemblers that are re-quired during the manufacturing process. In other words, variant assemblers have a workflow that can shift from high workload to low workload and vice versa in a short period of time. To solve this problem a prototype will be developed. This prototype will be used to check if it’s possible to optimize the work schedule for variant assem-blers with the help of predictive simulations. The result of the study became an implementation in form of a prototype. This prototype is built up in two layers; a data layer and a simulation layer. The data layer provides the simulation layer with two different datasets. The first dataset is based on historical data and is derived from Scania’s production in Zwolle. The second dataset is based on synthetic data which is formed with a high utilization rate in order to mimic a better production situation with less product variants to assemble. The simulation layer consists of a DES-model that is modelled after a station in the final assembly of Zwolle. After a simulation has been executed, this layer generates a simulation result in form of a graph that presents the utilization rate for a group of variant assemblers. This will happened for each dataset in the data layer, in this case two times. The simulation result that got produced shows that it’s possible to create a simulation with predictive characteristics. A long term solution for Scania’s problem statement requires more research within the possibility of combining different technologies such as DES with predictive methods such as ML and GAs. / Industriföretaget Scania CV AB är världsledande inom tillverkning av kommersiella fordon. De tillhandahåller ett modulärt system som inkluderar tunga lastbilar och bussar som kan konfigureras till en rad olika behov. Den här anpassningsförmågan leder dock till ett problem där varje order som tillverkas kan ha en stor varians av hur många montörer som krävs under produktion. I andra ord så har variantmontö-rer ett arbetsflöde som kan skifta från hög arbetsbelastning till låg arbetsbelastning och vice versa under en kort period. För att lösa dessa typer av problem så ska en prototyp med prediktiva egenskaper så som Diskrete Event Simulering (DES). Denna prototyp ska undersöka om det är möjlighet att optimera arbetsscheman för variantmontörer med hjälpa av prediktiva simuleringar. Resultatet av studien blev en implementation i form av en prototyp. Denna prototyp är uppbyggd i två lager; ett datalager samt ett simuleringslager. Datalagret tillhandahåller simuleringslagret med två dataset. Det första datasetet är baserad på historisk data och är härledd från Scania’s produktion i Zwolle. Det andra datasetet är baserat på syntetisk data som är framtagen med en högre utnyttjandegrad för att efterlikna ett bättre produktionssitation med färre produkt varianter att montera. Simuleringslagret består av en DES-model som är modulerad efter en station i slutmontering i Zwolle. Efter att en simulering har exekverats så genererar detta lager ett simuleringsresultat i form av en graf som presenterar utnyttjandegraden för en grupp med variant montörer. Detta sker för varje dataset i datalagret, i detta fall två gånger. Simuleringsresultatet som togs fram visar att det är möjligt att ha skapa simuleringar med prediktiva egenskaper. En långsiktig lösning för Scania’s problem-beskrivningen kräver mer forskning inom möjligheten att kombinera tekniker som DES med prediktiva metoder som ML och GAs.

Tillverkningsindustrin blir alltmer digitaliserad samt att nya digitala verktyg implementeras inom företagen. Som följd av detta pågår en förändring av arbetssätt. Smart Maintenance är det senaste begreppet i hur underhåll borde utföras inom tillverkningsanläggningar med hjälp av digital teknik. Detta begrepp syftar på ett arbetssätt som ämna möjliggöra en resurseffektivare produktion och underhållsverksamhet, ur såväl organisatoriskt som tekniskt perspektiv. I detta examensarbete genomfördes intervjuer med företag, vilket utgjorde den centrala undersökningsmetoden för att förstå hur den svenska tillverkningsindustrin ser på Smart Maintenance (SM), vad deras tolkning är på begreppet samt ifall de har tillämpat detta, samt tillämpat aspekter eller dimensioner från SM i deras underhållsverksamhet. En intervju med en forskare genomfördes för att utöka projektgruppens kompetens kring begreppet och dess påverkan på lönsamhet, hållbarhet och konkurrenskraft. Med information från intervjuerna och en litteraturstudie som grund, erhölls slutsatser kring vilka de främsta fördelarna och utmaningarna är i utövandet av Smart Maintenance, samt dessas samband med hållbarhet. Dessutom resulterade projektet i slutsatser kring hur företagen tolkar begreppet och hur data kan används för investeringsplaner inom de intervjuade företagen. / The manufacturing industry is becoming increasingly digital and new digital tools are being implemented within companies. As a result, there is a change in working methods. Smart Maintenance is the latest concept in how maintenance should be performed in manufacturing facilities using digital technology. This concept refers to a way of working that aims to enable a more resource-efficient production and maintenance operation, from both an organizational and technical perspective. In this thesis, interviews were conducted with companies, which constituted the central research method for understanding how the Swedish manufacturing industry views Smart Maintenance (SM), what their interpretation is of the concept and if they have applied this, and applied aspects or dimensions from SM in their maintenance operations. An interview with a researcher was conducted to expand the project group's knowledge on the concept and its impact on profitability, sustainability and competitiveness. Based on information from the interviews and a literature study, conclusions were obtained about what the main benefits and challenges are in the practice of Smart Maintenance, as well as their connection with sustainability. In addition, the project resulted in conclusions about how the companies interpret the concept and how data can be used in order to make better decisions within the interviewed companies.

Den moderna industrin har under senare år präglats av nya utvecklingar i teknikens värld, samt ett alltmer ökande behov av att effektivisera existerande processer och vara i framkant med innovationen av nya. Industri 4.0 är kulmen av detta samhällsbehov, en syn på industrin som till huvudsak drivs av cyberfysiska system, en integration av det fysiska med det virtuella i form av trådlösa uppkopplingar och molnteknologier. Internet of Things, även känd som IoT, är uppkopplingen av fysiska produkter till molnet som möjliggör datautvinning och övervakning under och efter produktionen i realtid. IoT förekommer i skiftande sammanhang och har tillämpats i varierande utsträckning inom olika branscher där det visat sig vara effektivt när det gäller inverkan på resursutnyttjandet. Inom handelsbranschen resulterade en IoT tillämpning på försörjningskedjan till en ökad kundkontakt och en förbättrad samarbetsrelation. Detta som en följd av reell-tids analyser av både behov, brister i försörjningskedjan och efterfrågan. Den har även en plats i moderna matbutiker som Electronic Shelf Labels (ESL). Detta tas reda på med hjälp av sökningar i journaler, konferensrapporter och ett teori baserad studium av tidigare tillämpningar. Målet med studien är att se hur tidigare anmärkningar förhåller sig till fordonsbranschen. Fordonsbranschen är i framkant vad gäller tillämpningar av industri 4.0 och IoT. Tillverkaren Scania är i begynnelsefasen av en eventuell storskalig övergång, och redan har tillämpningen fört med sig förbättringar. Dock återstår det mycket som måste uppklaras, bland annat sekretess angelägenheter och en omfattande kompetens kring ämnet. / In recent years, the modern industry has been characterized by new developments in the technological world. There is an ever-increasing need to streamline existing processes and a need to be at the forefront for the innovation of new ones. Industry 4.0 is the culmination of this need, a form of industry that is mainly driven by cyberphysical systems, an integration of the physical with the virtual in the form of wireless connections and cloud technologies. The Internet of Things, also known as IoT, is the connection of physical products to the cloud that enables data recovery and monitoring during and after production, in real-time. IoT exists in various contexts and has been applied to varying degrees in different industries, and has proven to be effective in terms of impacting resource efficiency. In the trading industry, an IoT application to the supply chain resulted in increased customer contact and an improved cooperative relationship. This is a result of real-time analysis of both needs, supply chain shortages and demand. It also has a place in modern grocery stores such as Electronic Shelf Labels (ESL). The information was sought after in online journals, conference reports and past applications. The goal of the study is to establish a relation between these past applications and the automotive industry, to find out how they compare. The automotive industry is at the forefront in terms of applications of Industry 4.0 and IoT. The manufacturer Scania is in the beginning phase of a possible large-scale transition, and the application has already brought improvements. However, much remains to be resolved, including confidentiality issues and extensive expertise on the subject.

The term ‘Industry 4.0’ leads to many new possibilities like smart factory which is the amalgamation of manufacturing systems in a network to perform tasks more efficiently. It is becoming more and more important for the companies to develop smart factories and integrate the devices within such a facility to meet the demands of the evolving market. The next generation production systems are designed to share the data within the network, plan, and predict the solution for the future problems. One such technology under smart factory is 3D laser scanning resulting in point cloud of the production unit. The traditional way of documenting a layout is usually with the help of 2D computer aided designs which are susceptible to measurement errors and changes that are not updated regularly. With the help of point clouds, an as-is representation of the factories can be recorded which can be easily updated with changes in the real world. With advancements in virtual manufacturing, the need for visualization of the factories is increasing drastically. 3D Laser Scanning is one of the better ways of meeting this need, among many other applications. The focus of the thesis had been to create a method document for 3D laser scanning of factories and to discuss the future possibilities of it. The research approach used in this thesis was conducting observational study, interviews and testing of the method. One such future possibility is autonomous scanning and how it would be beneficial for a company like Scania which is developing smart factories. Based on the study carried out during the thesis, a document presenting the method developed is included in the report. The report also points out the applications and benefits of point cloud over traditional layout planning methods.

IT is a well-integrated function within most companies and its importance grows bigger by the day. With new solutions and concepts being introduced continuously it is important to be aware of the ever-changing possibilities found within IT. One of these changes is the concept of Industrie 4.0 which poses as a revolutionary way to do business by connecting the real world with the virtual one to a greater extent than what is done today. Research has shown that there are many possible benefits of implementing Industrie 4.0 and also Smart Factory, such as improved inventory control and faster reaction time. Since these concepts are quite new, no real definition exists and the congruence between the academic and business world is not always at the same level, and therefore the first steps are not yet defined. Therefore, this study tried to reduce the gap between these two worlds by offering concrete recommendations of what needs to be done to be able to apply Industrie 4.0 in the real world at Scania CV AB and Scania IT. Scania CV AB posed as a case company to find out where to start on the road to become smart. Currently there are many functions using the services of Scania IT, but exactly how the systems are used is not known by Scania IT. To be able to provide the necessary services for the various functions of Scania CV AB and start the road of becoming smart, Scania IT needs to know how the systems are used and what information that is currently missing. A formulated strategy of Scania, as a whole, is to be able to collect and analyse information in order to have a more Intuitive Presence and Predictable Future, two words meaning that more proactive work can be conducted and more autonomous decisions can be made. To be able to fulfil this vision, knowledge about the needed information must be acquired by Scania IT. With focus on the information connected to the material flow before the material reaches the assembly lines found at Scania CV AB the purpose of this study was to identify and analyse information and actions needed in the material flow from a user perspective, to become Intuitive and Predictable as part of the concept Industrie 4.0. A set of research objectives were formulated as a guide for the study. By first identifying, with the help of the first research objective, the information input and output for the functions at Scania CV AB connected to the material flow, with a base in the functions planning material, it was identified that at different production sites different standards of working exist, but also differences in the IT usage and system configurations was found.  The second research objective focused on what information should be available for production and material planning according to a literature review and this was later compared with the findings at Scania, which composed the third research objective. As it turned out, Scania uses the correct set of basic information such as forecasting, production plan, and calculations of gross demand, along with information regarding costs, lead times, and inventory. However, how to use the information is not standardized and the users of the IT systems perceived the information as hard to find and difficult to interpret. The fourth research objective focused on the concept of Industrie 4.0 and Smart Factories by studying literature, an external company and the ideas that Scania CV AB have, to see what must be done before a Digital Factory can be created. The recommendations for Scania IT were based on the result on the analyses and they can be summarized by the need of further standardization of information and information usage to be able to start the road of becoming Smart and take one step closer to the concepts of Smart Factory and Industrie 4.0.

På Hannovermässan, världens ledande mässa inom industriell teknologi, i Tyskland år 2011 myntades begreppet Industrie 4.0 som beskriver den fjärde industriella revolutionen. Denna handlar om att integrera affärsprocesser med ingenjörsprocesser och på så sätt göra produktionen mer effektiv, miljövänlig och flexibel. Sedan 2011 har man även börjat tala om smarta fabriker och införande av smart teknik inom tillverkningsindustrin. Många hävdar att smart teknik och Industrie 4.0, tillsammans med vår tids krav på kundanpassade produkter, kommer att revolutionera den tillverkande industrin och förväntningarna är stora på vad smartteknik kan komma att bidra med. Denna studie syftar till att undersöka hur produktionen i tillverkande företag påverkas med hjälp av införande av smart teknik. Studien kommer att identifiera och analysera de främst påverkade faktorerna vid införande av smart teknik från år 2011 och framåt. De faktorer som identifierats och studeras närmare är; flexibilitet, kostnader, ledtider, miljöpåverkan, produktkvalitet och tillförlitlighet. I arbetet att hitta de faktorer som främst påverkar produktionen i tillverkande företag vid införande av smart teknik har främst två metoder använts. En inledande litteraturstudie gjordes för att identifiera de vanligast förekommande faktorerna. Valda faktorer verifierades därefter med hjälp av intervjuer och studiebesök hos två tillverkande företag som implementerat smart teknik, Scania CV AB och Volvo Car Corporation. Under vald tidsperiod har dessa företag tillsammans med KTH genomfört projektet FFI Line informationsystem architecture (LISA), där LISA är en variant av en smart fabrik. En slutsats som dragits i denna studie är att de faktorer som främst påverkas med hjälp av smart teknik inom tillverkningsindustrin är några av de som identifierades vid undersökningens start; flexibilitet, kostnader, miljöpåverkan och tillförlitlighet. Exempel på hur dessa påverkas presenteras vidare; om en informationsarkitektur (såsom LISA) implementeras blir det lättare att genomföra förändringar i maskinparken, det gör processkedjan mer dynamisk vilket ger en ökad flexibilitet för företagen. Med hjälp av uppkopplade enheter kan företagen snabbare informeras om fel i produktionen och på så sätt minska mängden kassationer vilket innebär kostnadsbesparingar. Genom att hastighetsanpassa motorer samt att optimera robotars rörelser, exempelvis genom att i större utsträckning använda mjuka rörelser, kan energi sparas och därmed ge en minskad miljöpåverkan. Dock blir företagen vid införande av smart teknik beroende av uppkoppling och det blir då viktigt att säkerställa att tillförlitliga system används om problem uppstår. Andra faktorer som påverkas är ledtider och produktkvalitet, som dock snarare påverkas som konsekvenser av att ovan nämnda faktorer påverkas. Ytterligare en slutsats är att Industrie 4.0 inom svensk tillverkningsindustri betraktas som en evolution snarare än en revolution och att den smarta tekniken är något som implementerats gradvis. / The term Industrie 4.0 was first coined in Germany 2011, at the Hannover Messe, the world’s leading Trade Fair for Industrial Technology. It is a term describing the fourth Industrial Revolution. This revolution concerns making production more efficient, environmentally friendly and flexible, by integrating business processes with engineering processes. Since 2011, smart factories and implementation of smart technology in manufacturing has also been a major talking point. Many claim that smart technology and Industrie 4.0, together with the increasing demands for customization, will revolutionize the manufacturing industry and the expectations for what smart technology will contribute with are great. The purpose of the study is to investigate what factors of the production in manufacturing companies are affected by the introduction of smart technology, from year 2011 onwards. The factors identified and investigated are; flexibility, costs, lead times, environmental impact, product quality and reliability. The study was conducted in two stages. First, a literature study to identify and investigate the factors, and second, visits, interviews and verification of the factors with two manufacturing companies who have both implemented smart technology; Scania CV AB and Volvo Car Corporation. During the selected time period, these companies have, together with KTH, participated in the FFI Line information system architecture project (LISA) project where LISA is a type of smart factory; an information architecture. A conclusion to be drawn is that the factors mostly affected by smart technology within the manufacturing industry are some of those identified in this study; flexibility, costs, environmental impact and reliability. Examples of impact are; if an information architecture (for example LISA) is implemented, it becomes easier to implement changes in the factory. This makes the process chain more dynamic, which gives the company greater flexibility. With the help of connected units, companies are faster informed about errors in the production, which makes it possible to decrease the amount of scrap. This leads to cost savings. Also, by using speed-controlled engines and by optimizing the movements of robots, energy can be saved, which leads to less environmental impact and also cost savings. Though, by implementing smart technology, companies also make themselves dependent on wireless connectivity. Therefore, it becomes important to make sure reliable systems are used. Other factors affected by the implementation of smart technology are lead times and product quality, but they are rather affected as consequences of the affection of the other factors, mentioned above. Another conclusion to be drawn is that Industrie 4.0, within Swedish manufacturing, is considered to be an evolution rather than a revolution and smart technology is something that is implemented gradually.